Keepalived HA配置和测试过程for linux.docx

Keepalived配置和测试过程for linux安装环境：两台server：System x3650 M3Node1：硬盘：130G CPU：2颗E5620 @ 2.40GHz 内存：4GNode2：硬盘：130G CPU：2颗E5620 @ 2.40GHz 内存：8G安装redhat 6.4系统用U盘安装rhel-server-6.4-x86_64-dvd.iso分区：/boot 200M/boot/efi 200MSwap 4096M/ 13000M只安装Basic版本，不要安装桌面需要在language里面选择chinese包安装系统完成后设置网卡eth0Node1：Node2：设置keepalivedNode1：! Configuration File for keepalivedglobal_defs { router_id 234567 #router_id，配置一个网络唯一的ID}vrrp_instance VI_3 { #为keeplived实例号，一般一台服务器只有1个实例，不用更改 state BACKUP #为keepalived初始状态，此处需要设置成BACKUP nopreempt #配置为非抢占模式，在node1切到node2，vip从node1飘到node2后，node1恢复后，服务不会再切回node1，而是继续让node2做MASTER，服务在node2上面继续跑 interface eth0 #为启用keepalived的接口，根据实际情况设置 virtual_router_id 40 #为keepalived虚拟ID，同一集群必须相同 priority 100 #服务优先级，MASTER设置为100，BACKUP设置为90 advert_int 1 #心跳区间设置为1秒 authentication { auth_type PASS auth_pass #设置验证密码，同一集群必须相同 } virtual_ipaddress { 20 dev eth0 label eth0:0 #设置服务IP，dev label可省略 } notify_master /etc/keepalived/notify.sh master #切换成master，执行一次脚本 notify_backup /etc/keepalived/notify.sh backup #切换成backup，执行一次脚本 notify_fault /etc/keepalived/notify.sh fault #切换成fault，执行一次脚本}Node2：! Configuration File for keepalivedglobal_defs { router_id 123456}vrrp_instance VI_3 { state BACKUP nopreempt interface eth0 virtual_router_id 40 priority 90 advert_int 1 authentication { auth_type PASS auth_pass} virtual_ipaddress { 20 dev eth0 label eth0:0 } notify_master /etc/keepalived/notify.sh master notify_backup /etc/keepalived/notify.sh backup notify_fault /etc/keepalived/notify.sh fault}设置notify. shNode1:#!/bin/bash#case $1 in master) service httpd restart #此行为应用启动脚本，根据应用更改 echo master /etc/keepalived/log ;; backup) service httpd stop #此行为应用启动脚本，根据应用更改 echo backup /etc/keepalived/log ;; fault) service httpd stop #此行为应用启动脚本，根据应用更改 echo fault /etc/keepalived/log ;;esacNode2:#!/bin/bash#case $1 in master) service httpd restart echo master /etc/keepalived/log ;; backup) service httpd stop echo backup /etc/keepalived/log ;; fault) serv